Harmonic mitigation placement within the system

[gprenger]

We are building a wastewater plant in regional Victoria Australia and I have a problem with harmonics mitigation. System particulars:
22KV Utility with 3050A available short circuit current.
One 22KV switchboard distributing power to four 2500KVA transformers.
Two 4000A main-tie-main switchboards(MSB1 and MSB2)distributing 415V Three phase 50HZ to six motor control centers.
MSB1 feeds four MCC's: Two are in the same room as MSB1 each with 500KVA VFD load and 300KVA linear load. Two other MCC's are 100m away each with 400KVA VFD load and 200KVA linear load
MSB2 feeds two MCC's both in the same room each with 1000KVA VFD load and 200KVA linear load.
Yes it is a harmonic rich environment.

Our specification states 3% THD at PCC and 5% THD at each MCC bus. We have specified 6 pulse VFD's with DC chokes.

My question is it better to install active filtration at each MCC bus or can we correct at the MSB level? My VFD vendor claims it will work better at the MSB level, but I am concerned about the harmonic currents within the MCC's.

Cheers!

 

[ozmosis]

gprenger
Are you outside the THVD limits your spec states? (I'm assuming a harmonic calculation has been carried out)

 

[gprenger]

Yes at the present we have active filters located at each MCC bus and our calculations indicate we are within specification. The VFD vendor claims we can still maintain specification on each bus even with filters on the MSB's. We are still in design and have not bought any equipment. We have ruled out individual passive filters, 18pulse drives, etc as too expensive. There are 136 drives ranging from .37kw to 260kw.

Cheers!

 

[ozmosis]

If you are using true active harmonic filters then these would be designed to handle the current of the specific harmonic current distortion rather than the overall rated load. So having them at the MCC or at the MSB's shouldn't make too much difference. The only thing I can think of is if you have redundancy within your MCC's per MSB then it makes little sense to have Active filtering per MCC as the overall THID at each MSB will therefore be less and if you have the filter per MCC then it will potentially be out of action.
I trust that makes sense as I seem to have gone into acronym overload.

 

[Skogsgurra]

More acronyming:

3 % THD at PCC is, IMHO, way too conservative. You can easily live with 5 % or more. Is your limit set by yourself, a consultant, a filter vendor or by the authorities?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ozmosis]

I would agree with Gunnar.
As you are in Australia, an installation like this would come under AS/NZS 61000-3-6 as your PCC is the 22kV network. As far as this standard is concerned, it would not apply to points within a consumers own network. The limits would therefore be defined by the consumer (consultant). Some consultants would typically use EN 50160 table 1 to define the specific limits (down the levels of each harmonic order) with a THD of 8%. This would normally be deemed acceptable.

 

[LionelHutz]

The real question is do you trust yourself or the VFD vendor?

I really wish people would specify if they are talking current or voltage distortion. I'm assuming your numbers are voltage distortion since that's what they are most likely to be.

If the feeders from MSB to MCC are low enough impedance then it is possible the active filter at the MSB would correct the voltage distortion out to the MCC. As an example, the active filter may correct the MSB voltage distortion to 1%. So, if the harmonic current flowing through the MCC feeders causes less than 2% voltage distortion across the feeder you are good.

 

[mls1]

Do the 2500kVA transformers have to be wye grounded on the secondary or can you diversify with a wye/delta, delta/delta, and a delta/wye? That would give you a net 18 pulse drive from three of your unit subs while using 6 pulse drives. You wouldn't be able to parallel transformers but then again with 2500kVA unit subs you're already getting up there in terms of fault current.

I agree with the above comments regarding focussing on the PCC. You also should consider the duration of the distortion. In the US we use IEEE 519 as the standard and it state:

". . .it is recognized that the stated current distortion limits can be exceeded for periods of time without causing harm to equipment. When evaluating user compliance with the stated limits, it is recommended that probability distribution plots be developed from the recorded data and analyzed. If the limits are only exceeded for a short period of time, the condition could be considered acceptable."